Jensen PI-XX Isolation Transformer Review

[MaxwellsEq]

What a piece of snake oil scam!


It only changes the signal in audible and if you look carefully it actually adds distortion!

Spoiler: spoiler

This obviously is sarcasm but nevertheless I see double standards here.
With cables and power conditioners, other products get instantly labeled as snake oil without a fair test
(they can’t fix a problem that's not there).


Same here, but this somehow gets a pass because it could conceptually work?
If CMRR of any particular AMP or DAC/AMP is never measured, how can we know if this improves CMRR and how much?
And how much would it change with using better cables or maybe a USB "filter" or isolator?

When I started working in recording studios, transformers of this nature were used a lot. Some very expensive mixing consoles had one per channel built in. There would also be a cupboard with several standalone transformers. USB was years away from being invented.

Time is money and engineers can't waste time trying to find what's causing a hum or buzz. Bands being in their preferred keyboards and processors, many of which are single-ended or badly balanced with questionable behaviour. These sorts of transformers were used to make such devices "safe" to plug into the very expensive mixing console.

The conclusion you might want to draw is that a great deal of older (pre 1990s) recorded music you have heard has passed through at least one of these sorts of devices.

 

[torgeirs]

So where is the data and the test for this?

https://www.analog.com/media/en/training-seminars/tutorials/MT-042.pdf

+ if mains voltage get into the ground or signal the transformator will protect you.
The other thing the transformers do is blocking DC. They must be desaturated, but you will protect your speakers if the DC blocking caps further up your chain is malfunctioning.

Agree that it is snake oil if sold as sound enhancing device on modern low voltage equipment.
Its just a problem solver or life jacket.

 

[torgeirs]

In the olden days analog brodcasting studios equipment wasn't considered PRO if they did not have transformators in input/output. In contrary to snake oil you could not work professionally without it.

 

[Lambda]

When I started working in recording studios, transformers of this nature were used a lot. Some very expensive mixing consoles had one per channel built in.

Some equipment still has. Some amps have transformers at the input, some DAC have transformer output.
There are other ways, electronic ways, to make an input and output behave like it’s transformer-driven (within limits).
And there are other ways to isolate like USB and Toslink...
CMRR is very much dependent on the combination of source and sink (like DAC and AMP).
This can potentially fix a problem that would not be there with proper equipment.
There is just no excuse for an expensive DAC to share analog and digital ground, but many do and maybe have good SINAD if CMRR is ignored while testing.

+ if mains voltage get into the ground or signal the transformator will protect you.

And toslink, or a USB isolator, or spdif...

In the olden days analog brodcasting studios equipment wasn't considered PRO if they did not have transformators in input/output. In contrary to snake oil you could not work professionally without it.

Too bad almost nothing here gets tested if it has a transformer, DC protection, proper balancing inputs and outputs, or good CMRR...
There are still source and sink devices with transformers or other (functional) isolation on the input/output.
But if we would not know
and looking at a DAC with this transformer built in it would be criticized for its bad SINAD performance and distortion.
This is the double standard...
Sure this can help in some cases.
(so can potentially good cables vs bad cables, power conditioner/isolation transformer, "USB filter" and other quackery...)


But in these cases it’s always dismissed with
"Well the problem should not be there in the first place..."
"With good proper engineered equipment this would not be a problem"
"Why put lipstick on a pig"

 

[somebodyelse]

So where is the data and the test for this?

See for example https://www.ti.com/lit/an/sboa582/sboa582.pdf for the resistor tolerance component of CMRR in classic differential arrangements. In theory you could use precision matched resistors in both output and input to get better than 60dB CMRR, but in most commercial equipment that's unlikely. Bill Whitlock's U.S. Patent No. 5,568,561 improves this to ~90dB without needing precision matching, currently exclusive to THAT's 1200 series balanced line receivers I think.

"Would it not be be better if the Equipment would not need this. "
"With porperly dseinged equitment you would not need this"
This or something similar is the argument Argument given with USB filters and other devices.

In the pro world, and to the sensible hobbyist, they're just tools to be used where the situation requires them. In an ideal world we wouldn't usually need such isolators. Unfortunately we don't live in an ideal world - some manufacturers still don't follow AES48 and it's not usually obvious until you run into a problem. If the equipment is in your control, and you have the time, you could replace it or fix the problem, but in the pro world these aren't always true. If you're lucky a ground lift switch will fix it. If not you need something else, so having a transformer isolator in your kit bag is a good fallback. It's a tool to get a job done in a limited time with limited equipment, not something you would usually plan on using. I say usually based on instrumentation experience, because occasionally there are unusual situations when you absolutely do need galvanic isolation. An alternative to an audio transformer would be ADC -> isolated digital interconnect -> DAC if you needed better performance at the cost of some convenience.

 

[torgeirs]

And toslink, or a USB isolator, or spdif...

Agree that it is best solved with galvanic or opical isolation in digital domain. Then the transformers or modules can be much smaller and don't get LF that saturates them.
But when the user is in analog domain only, the line transformator can be handy.

But guess its most handy with a life jacket on open ocean, not so relevant at your swimming pool

 

[KSTR]

Transformers are very valuable in situations where you need a simple and quick fix with 100% guarantee that it will work, like on stage and in the studio. The increased LF distortion is irrelevant except for very rare cases.
Also, sometimes you have conditions that really cannot be solved with an analog balanced input, like too large common-voltage and/or required very high common-mode impedance. The ADC-->optical-->DAC idea could help there if the power supplies of the sides are adequate for the problem, which sometimes means battery supply or very specialized mains supply with excellent isolation properties. For this solution to be as good as a simple transformer the cost is significant.

 

[pma]

Optical fibre is by far the best solution (I do have some experience in isolating hundreds of kV in HV voltage testing labs). Audio transformers sometimes fail (in audio, of course), specifically when both source and receiver parts are floating and still cannot be connected by a wire.

 

[EERecordist]

Very good information on frequency and distortion. I have never dug to that depth. I'm sure some ASR participants know more than me. And it is likely Jensen engineers would enjoy a note.

Speakers have LF distortion, so I guess the transformers and speakers can compete for bad.

Transformers were necessary for tube amplification to isolate high voltages from the signal in the very early days. In the following solid state days, pre-about 1980, transformers were everywhere in professional audio. They were useful in solid state of that era for (balanced) common mode noise rejection. One of the early innovators was Neil Muncy with his Suburban Sound transformer-coupled op amp microphone mixers. In small signal applications you can use them for no-noise voltage gain.

The typical recording signal path in the 1970s would be microphone output transformer> microphone preamp input transformer> microphone preamp output transformer > any effects/echo/processing outs and in + the outboard equipment in and out transformers> possibly mix bus in and out transformers> multitrack recording console output stage in and out transformers> Dolby in and out transformers> tape machine in and out transformers. Then the music would come back from the multitrack tape for mixing. That would duplicate the same path because the multitrack would come as line in through transformers on the console preamps. Then the stereo mix would go out through the Dolby + tape transformer chain. The tape would be sent to mastering with their tape machine and Dolby transformers, mastering console and outboard compression and EQ, all transformer in and out at each stage, then to the vinyl cutting chain with transformers or to the A to D for CDs.

With op amps, it became possible to do common mode noise rejection without transformers. Capacitive signal coupling also went out of style.

Today, transformers are very common in professional microphones, feeding microphone preamps with or without transformers. That is because with millivolt microphone signals, common mode noise rejection is important. A small number of professional microphones will have transformerless balanced out. Mostly digital recording chains end-to-end are more possible today if the delay works for what you are doing.

Audio transformer manufacturing today is hard to find, and detailed technical specs even harder. Glad Jensen is continuing. They might be good to interview or write a history of transformers and what they do for ASR.

(I had many hard copy large format console blueprint schematics I threw out when I went to the dark side of chips and digital which would be valuable today, like all the vintage audio gear I gave away!)

 

[alebucc91]

Not necessarily, unfortunately. AES-48 has said how balanced connections should be implemented for many years now, but still some manufacturers don't follow it. This means that there are still products on the market with a 'Pin 1 Problem' - in other words still possibly causing ground loop issues even with a balanced connection. Any manufacturer not following the standard really deserves naming and shaming.

This device, reviewed by amirm, is useful for solving audible problems, right? It's not an "audiophile refinement."

 

[Dougey_Jones]

@amirm I still need to send you my Rolls "Buzz Off" for testing, it definitely eliminates hum/buzz like it claims, but I'm curious if it degrades sound. I currently use it between my x3700 and my Apollon amps because my HTPC introduces HF noise when connected to the HTPC via HDMI. What's weird is that the Denon is able to isolate and get rid of the noise without the help of the buzz off if I'm using it's internal amplifiers, but for some reason, when I use pre-outs the noise is present..

 

[somebodyelse]

This device, reviewed by amirm, is useful for solving audible problems, right? It's not an "audiophile refinement."

Yes - this device is a tool for providing isolation when necessary, such as when you have a 'Pin 1 Problem' that you can't solve in a better way (like replacing or modifying the offending equipment). There's a detailed description of the process for tracking down the offending equipment, or testing equipment to see whether it has a 'Pin 1 Problem' in Jensen's app note AN007 and a few of their others. Most of the buyers will be audio professionals who know what it is and how and when to use it. There are probably some audiophiles who use it too, either for its intended purpose (say they have something that has a problem, but that they don't want to part with) or perhaps as a 'refinement' - maybe they love transformers like some love tubes/valves?

 

[IamJF]

maybe they love transformers like some love tubes/valves?

Transformers are highly liked in e.g. mic preamps to get some sound colours. They are sometimes extra build with an output pad so you can drive the transformers to get the sweetspot for the right sound for your source.
I use Softunbe Console 1 for that use and depending on the music it's surprising how much "distortion" is used/needed to get it right.

 

[Dougey_Jones]

This device, reviewed by amirm, is useful for solving audible problems, right? It's not an "audiophile refinement."

Correct, this is for fixing a problem that exists, its not additive.

 

[Jimster480]

Thanks for your reply, but then I need more explanations.
Where to place a volume pot then? Upstream or downstream?
Why couldn't it be integrated to the device?

Downstream I think would be the way to do it.